MDM Scorecards
The MDM Scorecards page at /mdm/scorecards measures master-data health: completeness, accuracy, freshness, uniqueness, and conformity per entity type. It's the DQ Scorecard's sibling — same layout, same letter grades, but the metrics are MDM-specific.
Dimensions measured
| Dimension | Definition |
|---|---|
| Completeness | % of required attributes populated |
| Accuracy | % of attributes passing validation rules (DQ Suite) |
| Freshness | % of records updated within their SLA window |
| Uniqueness | 1 - (duplicates / total) — measures consolidation effectiveness |
| Conformity | % of attribute values matching reference data / patterns |
A weighted average produces the score. Default weights (overridable):
- Completeness 25%
- Accuracy 30%
- Freshness 15%
- Uniqueness 20%
- Conformity 10%
Views
- By entity type — Customer / Product / Account
- By source — how clean is each source system feeding MDM
- By steward — quality on entities owned by each steward
- By domain — for orgs that scope MDM by business domain
Targets and watchlists
Set per-entity-type target grades. Below-target cells highlight red and feed the Watchlist strip.
Drill-in
Click any cell to open detail showing:
- Which attribute(s) are dragging the score
- Sample failing records (top 100)
- Change in score over the past 30 days
This is where you diagnose why score dropped — was it a new source feed dumping malformed data? A stewardship backlog? A retired ref-data code that hasn't been cleaned up?
Coverage gap
Special view: Uncovered entities — entities with no DQ rules attached. Encourages building rule coverage so the score is meaningful.
Sharing
Export as PDF or PNG for monthly reviews. Snapshot to Delta for trend audits.
API
GET /mdm/scorecards?view=by_entity_type
GET /mdm/scorecards/{entity_type}/history
GET /mdm/scorecards/{entity_type}/failing-records
Related
- MDM Profiling — drill into attribute distributions
- DQ Scorecard — sibling for non-MDM data
- MDM Reports — exportable detail